Styles are trained on synthetic data and do not copy the styles of artists!
Photo 2 - is trained on real data (Dataset from Unsplash)
[M] Merged LoRAs
Recommended LoRA Strength (Weight): 0.8-1
Preview generated without using ADetailer & Hires. fix
Description
Repeats: 20
Epoch: 6
Image Count: 24
Network Rank: 32
Network Alpha: 16
Batch size: 2
Clip skip: 1
Training steps: 1440
Resolution: 832x1024+1024x768
Train Resolution: 1024x1024
Model: ponyDiffusionV6XL_v6
WD14 Captioning
SmilingWolf/wd-v1-4-moat-tagger-v2
LoRA Weight: [0.7-1]
This Style is trained on the images created by this LoRA
FAQ
Comments (12)
There Pony XL Loras are invaluable, thank you for your contribution.
These make producing great images with PonyXL so much easier. While I love PonyXL, it can be difficult to work with at times. My gens have improved greatly since I started using these LoRAs. Thank you.
PS. One of my favorite ways to use them is to combine concept twilight with gothic and either anime 2 or line art. Good stuff. 👍
Thanks, glad you like it. ❤️ I stopped using other models except Pony Diffusion V6 XL, what I like most about this model is that you can create SFW/NSFW content with men. most SDXL models are good at creating only women, this is the first model that is good at creating men and women equally
@prgfrg23 I completely agree. 😁 I have used an SD1.5 model once (Dark Sushi for a landscape of a demon lord's castle at night) since I started using PonyXL. Until I got Pony, I had no hope for generating nsfw on SDXL. And I also generate a lot of guys. Pony truly has very little bias with regard to male/female content, and it seems like the creators have tried to keep everything else fairly even and unbiased too, although it does seem be a bit heavier on cartoon-style than anime-style data.
There were so many things I tried making when I first started into Stable Diffusion, and nothing could do it. But Pony is making a lot of those things possible for the first time. It's really reenergized the stuff I create. Heck, I've even gotten it to generate a 69 position several times.
One of the big parts of Pony working so well is your LoRAs, and that's not just flattery. I don't use (or haven't been, at least) any other LoRAs with Pony. Only yours. They work so well together and with Pony that between everything Pony can do and your LoRAs, I just haven't needed any others.
I'll post some of my recent gens in the gallery. I think you'll like them. Or, I hope you will anyway! We probably have different tastes in guys (mine seems to be different from almost all the other gay/bi guys here) but I hope you find them enjoyable anyway. For now, though, it's bedtime. I'll try to post those tomorrow.
why are some strengths recommended at 1 but others a bit less?
Some models have a very strong style
minimum weight 0.7 maximum weight 1
If the weight is less than 0.5, the style is almost invisible
sometimes even at weight 1 the style can change
My LoRAs do not create a specific style but just control the model making the style generation more stable.
Can't wait for Photo V2, wishing you all the smooth progress possible.
Has it recently been changed to Pony from SDXL?
Cause it worked like two days ago on mage.space. Which prevents us from importing it there. :/
Do I understand correctly that models marked as Pony don't work on mage.space?
If so I can change everything back to SDXL if that helps.
I apologize if this change to Pony has caused any inconvenience.
Is there a Pony Diffusion V6 XL model on mage.space or not?
If mage.space doesn't have this model then it doesn't make any sense, These LoRA's are only for the Pony Diffusion V6 XL
The license of Pony Diffusion v6 prevents commercial inference, so it wouldn't be legal for mage.space to run Pony Diffusion anyway.
However, you can generate images with Pony Diffusion and this LoRA for free here on Civit since a few days ago. Civit has an explicit permission from PD authors for commercial inference.
@prgfrg23 Sorry for late reply, I appreciate your request to change back. But I bet I am part of a minority here. So don't worry about that.
PonyXL V6 is marked as 'SDXL 1.0' Base Model on Civit, So it can be imported and used there. But if it changes to 'Pony' Base Model, it will no longer be importable.
It's a Mage.Space problem, not civit or their user's fault. I'm sure I can bypass this by downloading your file and then upload it to a google drive and then import it to Mage.Space, cause then I have to manually select if it's 1.5 / 2.1 / SDXL 1.0.
Cause importing from civit.ai automatically use what is already listed there. And Mage.Space only has 1.5/2.1/SDXL, not Pony ^^
(Sorry for my rambling explanation. Hope you got the gist of what I meant to convey <3)
You changed it back! <3 <3 <3
Details
Files
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Mirrors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
hf_model_48_16.safetensors
Digital Art Style.safetensors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Digital Art Style SDXL_LoRA_Pony Diffusion V6 XL.safetensors
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.






